[LU-17320] sanity-flr test_200: read failed (-ESTALE)

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Affects Version/s: Lustre 2.15.5
• Severity: 3
• Assignee: Fred Dilger
• Reporter: Maloo

Description

This issue was created by maloo for eaujames <[e-mail]>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/6fa7c859-c71d-4ccb-b841-c10d6d95620b

test_200 failed with the following error:

read failed

Test session details:
clients: https://build.whamcloud.com/job/lustre-reviews/100514 - 4.18.0-477.15.1.el8_8.x86_64
servers: https://build.whamcloud.com/job/lustre-reviews/100514 - 4.18.0-477.15.1.el8_lustre.x86_64

dmesg client:

[ 5869.288546] Lustre: DEBUG MARKER: /usr/sbin/lctl mark == sanity-flr test 200: stress test ====================== 14:21:02 \(1701094862\)
[ 5869.497763] Lustre: DEBUG MARKER: == sanity-flr test 200: stress test ====================== 14:21:02 (1701094862)
[ 5869.573523] Lustre: DEBUG MARKER: mkdir -p /mnt/lustre2
[ 5869.587901] Lustre: DEBUG MARKER: mount -t lustre -o user_xattr,flock trevis-30vm6@tcp:/lustre /mnt/lustre2
[ 5869.639207] Lustre: Mounted lustre-client
[ 5869.674845] Lustre: DEBUG MARKER: mkdir -p /mnt/lustre3
[ 5869.687743] Lustre: DEBUG MARKER: mount -t lustre -o user_xattr,flock trevis-30vm6@tcp:/lustre /mnt/lustre3
[ 5925.096111] LustreError: 353348:0:(lcommon_cl.c:196:cl_file_inode_init()) lustre: failed to initialize cl_object [0x20000716a:0x2d:0x0]: rc = -116
[ 5925.366809] Lustre: DEBUG MARKER: /usr/sbin/lctl mark  sanity-flr test_200: @@@@@@ FAIL: read failed 
[ 5925.581296] Lustre: DEBUG MARKER: sanity-flr test_200: @@@@@@ FAIL: read failed

This issue seems to appear mostly on b2_15 or in interop testing:
https://testing.whamcloud.com/test_sets/27129ab9-da73-44e5-98bc-b341565b157b
https://testing.whamcloud.com/test_sets/962c363d-ee97-415c-a3cc-97d088e23d3d
https://testing.whamcloud.com/test_sets/5467efe4-b33b-4068-9fed-dccf9869d1a0
https://testing.whamcloud.com/test_sets/171a2784-e312-40c2-995d-78a7ec5dff24
https://testing.whamcloud.com/test_sets/263942a5-e3bf-47fe-a0f8-513c628657ad
https://testing.whamcloud.com/test_sets/be383e45-fad1-417f-a711-a656fe2f4446
https://testing.whamcloud.com/test_sets/da65b1f5-49e0-460b-b1d2-d6504c7a6aaf

VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV
sanity-flr test_200 - read failed

Issue Links

• duplicates: LU-15300 mirror resync can cause EIO to unrelated applications

Comments

[Fred Dilger]

LU-15300 backported and merged to b2_15 branch.

[Peter Jones]

Fred is porting the LU-15300 patch

[Minh Diep]

Failure Rate: 16.00% of most recent 100 runs, 0 skipped (all branches)

[Sarah Liu]

+1 on 2.15.5 https://testing.whamcloud.com/test_sets/5f336f6d-0820-431c-9a82-4d6f1d9efaf5

[Alex Zhuravlev]

probably LU-15300 patch is missing on the server?